Title: MM-wave double drift IMPATT diode: a comparison of experimental, theoretical and neural network approach

Authors: P.R. Tripathy; S.K. Choudhury; S.P. Pati

Addresses: Department of Electronics and Communication Engineering, Gandhi Engineering College Bhubaneswar, Gandhi Vihar, Badaraghunathpur, Madanpur, Bhubaneswar-752054, Odisha, India ' Department of Physics, Gangadhar Meher University, Fatak, Budharaja, Sambalpur, Odisha, Pin: 768004, India ' Department of Electronics and Communication Engineering, National Institute of Science and Technology, Palur Hills, Berhampur, Odisha, Pin: 761008, India

Abstract: This paper presents the comparison of simulation and experimental output with the neural network results of different materials like Si and GaAs-based IMPATT diode at 94 GHz window frequency. The experimental results are nearly equal to simulation results and show a quantitative agreement between theoretical values and experimental output. The experimental results in terms of efficiency and RF power are almost 80-85% of the simulation results. Rest has 15-20% discrepancy between theory and experiment due to several causes like temperature, small-signal approach, parasitic effects, impedance matching problem and proper heat-sink arrangements. The developed neural network model is used to optimise performance of the diode by taking other wide band gap semiconductor material-based IMPATT diode. The model and the results will be very useful for applications in radar and guided missile technology.

Keywords: silicon; gallium arsenide; double drift; IMPATT; RF power; efficiency; neural network.

DOI: 10.1504/IJNBM.2018.094902

International Journal of Nano and Biomaterials, 2018 Vol.7 No.4, pp.262 - 274

Received: 28 Apr 2017
Accepted: 04 Dec 2017

Published online: 26 Sep 2018 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article